Tidy Data
Presenter: Olivia Beck
Content Credit: Matthew Beckman, Hadley
Wickham
May 17, 2023
Tidy Data
“Happy families are all alike; every unhappy family is unhappy in its
own way.” –– Leo Tolstoy
“Tidy datasets are all alike, but every messy dataset is messy in its
own way.” –– Hadley Wickham
- Key ideas:
- Cases = Rows
- Variables = Columns
- How should we define case?
- How do we identify variables?
- Advantages and Disadvantages
What is Tidy Data
- Being neat is not what makes data tidy!
There are three interrelated rules which make a dataset tidy:
- Each variable must have its own column.
- Each observation/case must have its own row.
- Each value must have its own cell.
It is your job as the researcher to define the variables,
observations, and values.
- The “tidyness” of the data set depends on the research question. It
is not an inherent property to the data set itself.
- When data are in tidy form, it’s often straightforward to transform
the data into arrangements that are useful for answering interesting
questions.
Example of Untidy data
Example of Tidy Data
- Disadvantages
- tidy data can be hard for human to quickly interpret
- often not the ideal form for creating graphics
- Advantages
- clear definitions
- tidy data can easily be wrangled to a useful form for
interpretation and visualization
Galton Data
In the 1880s, Francis Galton started to make a mathematical theory of
evolution.
Here’s part of a page from his lab notebook. Discuss the following in
groups:
- What might he investigate with these data (e.g., Research
Question)?
- Are these data tidy according to our
definition?
- What are the cases?
- What are the variables?
- How many rows of data should the result have?
- How many columns of data should the result have?
What is the data type of each column?
- What are some additional variables (not yet shown) that might be of
interest? How would you recommend showing that information in the data
table?
